MEAL-KITS: WHAT DRIVES CONSUMPTION?
Master Thesis Marketing
Tom van Schijndel S4485068
Radboud University Nijmegen, 2020
Master thesis – Marketing
Supervisor: dr. M. Hermans
Meal-kits: What drives consumption?
A research study on the drivers for the consumption of meal-kits, the difference
between specialized providers and grocery stores and the influence of brand
loyalty.
Master’s thesis Marketing (MAN-MMATH-2019-FM-JAAR-V)
Tom van Schijndel (S4485068)
Counselors:
dr. M. Hermans
I proudly present to you my master’s thesis in marketing “Meal-kits: What drives
consumption?”. This thesis investigates some drivers for the consumption of meal-kits by consumers, explores the difference between specialized providers and grocery stores offering meal-kits while looking at the moderating effect of brand loyalty. This topic is chosen because I was curious about the topic and interested to learn more about it. Completing this thesis will contribute to finishing my master’s degree at the Radboud University Nijmegen, with a specialization in Marketing. The marketing program started in September 2019 and ended August 2020.
This research builds on authors like Brunner, van der Horst and Siegrist (2010) who investigated the drivers for the consumption of convenience food. Meal-kits are a special form of convenience food and this topic has not received sufficient attention in the literature. This thesis investigates some of the most important drivers found multiple studies and applies them to the context of meal-kits. Furthermore, the difference in drivers between specialized
providers and grocery stores is a central aspect of this study as well as the moderating effect of brand loyalty.
During the process of writing my thesis I started in a group of seven students and with Ms Hermans as our supervisor. There were three main topics in our group and after some group meetings the group was divided into three according to the subject you had chosen. The meetings were pleasant and the atmosphere was always contributing to overcome any
obstacles. While writing my thesis I encountered multiple obstacles and sometimes felt a bit lost. But after consulting with Ms Hermans or other students it worked out at the end. Therefore, I would like to specially thanks Ms Hermans for the total guidance process even when this was made a bit trickier because of the corona virus. Moreover, she made the whole process of writing a master’s thesis a pleasant and informative experience.
Enjoy reading,
Tom van Schijndel
Meal-kits are becoming increasingly popular in Western society and this is reflected by the vast amount of meal-kit delivery services and grocery stores selling meal-kits. Western people are preoccupied with their job and perceive their life to be very busy. To decrease the time and effort involved with cooking a meal, the need for convenient meals is increasing. Prior work in the field of convenient meals is predominantly focussed on what exactly is
convenience food and what drives the consumption and purchase intention of it. Drivers such as price, effort reduction and healthiness are identified as important factors in determining the consumption of convenience food. So far, little is known about what drives consumers to buy meal-kits and whether the drivers for convenience food are also important for the purchase intention of meal-kits. Further, the difference in drivers between specialized providers and grocery stores as well as the role of brand loyalty remains unclear.
This study verifies if the drivers for the consumption of convenience food are also important for meal-kits while accounting for the moderating effect of specialized providers versus grocery stores and brand loyalty. The results show that price, effort reduction and healthiness do not affect the consumption of meal-kits. Moreover, there is no difference in drivers between specialized providers and grocery stores offering meal-kits. Furthermore, brand loyalty does not affect the relationships between the drivers and the purchase intention but does affect the purchase intention directly. These findings help managers and marketers in developing more knowledge on what factors are not important when they want to increase the purchase intention of their customers.
1. Introduction 1 1.1 Meal-kits 1 1.2 Research aim 2 1.3 Research question 2 1.4 Theoretical relevance 2 1.5 Practical relevance 4 1.6 Outline 5 2. Theoretical Framework 7 2.1 Introduction 7 2.2 Convenience food 7 2.2.1 Meal-kits 10 2.2.2 Purchase intention 12 2.2.3 Brand loyalty 13 2.3 Conceptual model 14 2.4 Hypotheses 15 2.4.1 Price 15 2.4.2 Effort reduction 16 2.4.3 Healthiness 17
2.4.4 Specialized providers versus grocery stores 18
2.4.5 Brand loyalty 21 3. Method 23 3.1 Introduction 23 3.2 Research design 23 3.3 Procedure 24 3.3.1 Survey 24 3.4 Research ethics 26 3.5 Measures 26 3.5.1 Purchase intention 27
3.5.2 Specialized providers versus grocery stores 27
3.5.3 Price 28
3.5.4 Effort reduction 28
3.6 Analysis 30 4. Results 34 4.1 Sample 34 4.2 Data analysis 35 4.3 Robustness checks 39 5. Discussion 41 5.1 Theoretical contributions 41 5.2 Practical implications 45
5.3 Limitations and future research 46
References 49
Appendixes 53
Appendix I: 53
Appendix II: 64
1. Introduction
The retailing industry is dynamic in nature and is increasingly being influenced by rapid technological changes and a changing competitive market structure (Kumar et al. 2017). As a result, retailers are trying to come up with new business models and practices to keep up with the changes and demands in this competitive environment. One important innovation in retailing and specifically the food industry are meal-kits (Shankar & Yadav, 2011).
1.1 Meal-kits
One of the developments in consumer behaviour is the increasing need for convenience (Brown & McEnally, 1992). Meal-kits are a way to meet this demand by retailers. Meal-kits are packages containing the right amount of ingredients and a detailed instruction on how to make a complete meal. By introducing meal-kits, task formerly done by the consumer are now offloaded to the producer which results in more convenience for the consumer (Inman, & Nikolova, 2017). Because specific tasks such as coming up with a recipe are now offloaded to the producer, meal-kits can be regarded as a form of convenience food (Jackson and Viehoff, 2015). When buying a meal-kit the consumer does not have to come up with recipes themselves and has a stepwise instruction to follow when preparing the meal, so purchasing a meal-kit saves time and effort (Jackson and Viehoff, 2015). Consumer research conducted by Friends and Goldin (2019) surveying more than 10.000 American Households, shows that the main reason for ordering meal-kits is convenience which is reflected by the amount of time and effort saved by cooking with meal-kits.
Meal-kit services where you can order a meal-kit online and have it delivered are becoming very popular with companies such as HelloFresh and BlueApron dominating this market (Financial Times, 2017). In 2017 the total meal-kit industry is valued around $2.2 billion and is expected to rise to over 6 billion in three years (Friends and Goldin, 2019). The annual growth rates for the meal-kit industry are forecasted at 25% to 30% for the coming 5 years (2017 to 2022) making it a promising area for development and further research (Friends and Goldin, 2019). Moreover, consumer research shows that 3.8% of the surveyed households ordered a meal-kit in the past 30 days and 27% says that they are interested in buying a meal-kit in the coming 30 days (Friends and Goldin, 2019). In the US alone
approximately 150 meal-kit providers came to life from 2011 to 2016 (PYMNTS, 2019). With online and traditional retailers entering the market the bigger corporations such as Amazon
are also investing in the meal-kit market. For example, Amazons acquisition of Whole Foods to launch a meal-kit delivery service in the US (Amazon, 2019). Besides that, major food companies such as Nestlé and Unilever are investing in meal-kit providers (PYMNTS, 2019).
Because of the popularity, enormous growth potential and the potential threat they pose to traditional grocery retailers such as AlbertHeijn and Jumbo, these retailers are introducing their own meal-kits to compete against companies such a HelloFresh
(AlbertHeijn, 2019). Thus, there are two types of meal-kits which deviate from another based on the provider offering the meal-kit. Meal-kits from specialized providers are offered online and are delivered to your address often on a subscription basis and meal-kits from grocery stores can be purchased in the physical store without subscription.
1.2 Research aim
The goal of this research is to verify if the drivers price, effort reduction and healthiness which are suggested to be important drivers for the consumption of convenience foods, (Brunner, 2010 Hertz and Halkier, 2017; McIntosh, 1996; Rappoport, et al., 1993) are also important for meal-kits. Additionally, two kind of providers, specialized providers and
grocery stores, are compared based on those three drivers. Furthermore, the moderating effect of brand loyalty is taken into account to study whether varying levels of brand loyalty causes the drivers to be different.
1.3 Research question
The research question central in this thesis is: To what extent are price, effort reduction and
healthiness drivers for the purchase intention of meal-kits?
Sub questions:
- To what extent does the difference between specialized providers and grocery
stores moderate the relationship between price, effort reduction and healthiness on the purchase intention of meal-kits?
- To what extent does brand loyalty moderate the relationship between price, effort
reduction and healthiness on the purchase intention of meal-kits?
In the literature on convenience food the concept of meal-kits is relatively
understudied (Jackson & Viehoff, 2016). Despite the promising potential of the overall meal-kit market and the fierce competition between providers, knowledge about the subject is scarce. The problem central to this study is that it is not known if the drivers for the consumption of convenience food also apply to meal-kits. When reviewing the literature multiple drivers for the consumption of convenience food are identified but it is not yet empirically verified if these drivers are also applicable to the consumption of meal-kits (Brunner, 2010; Hertz and Halkier, 2017). Meal-kits reduce time and effort involved with coming up with recipes and shopping for ingredients and are therefore often considered as convenience food. This is not correct because meal-kits still need mental and physical effort to cook the meal. Meal-kits are also more expensive and healthier compared to traditional convenience foods (Nielsen, 2017; Moorhead, 2016). This deviation from convenience foods is a reason to believe that the drivers for convenience food do not apply to meal-kits. There are many important drivers for the consumption of convenience food (Brunner, 2010). The drivers central in this study are price, effort reduction and healthiness because these are the aspects that cause meal-kits to be different from traditional convenience foods. The choice for these drivers is based on the literature on convenience food choice (McIntosh, 1996;
Rappoport et al., 1993; Steptoe et al., 1996; Brunner et al., 2010). The first contribution is focussed on verifying if the difference in underlying characteristics for meal-kits causes the drivers to deviate in comparison with convenience food in general.
The introduction of meal-kits in grocery stores is a reaction on the success of the specialized meal-kit providers. The concept of the meal-kits provided by two different producers is the same but the perceptions from consumers about both varies (Nielsen, 2017; Moorhead, 2016). Some authors argue that meal-kits from specialized producers are perceived to be of higher quality, more varied, healthier and give a better ‘eating-out’ experience
(Nielsen, 2017; Moorhead, 2016). In addition, they argue that meal-kits from specialized providers are perceived as more expensive and demand more effort than meal-kits from grocery stores (Nielsen, 2017; Moorhead, 2016). The difference in perceptions of consumers about the meal-kits from specialized meal-kit providers versus those offered by grocery stores is not empirically verified (Hertz & Halkier, 2017). The effect of this difference in perception on the drivers price, effort reduction and healthiness is also not empirically verified. This is important because meal-kits cannot be understood as one concept. The provider that offers the meal-kit determines the configuration and the aspects such as price, effort reduction and healthiness. The most important theoretical distinction that must be made is that between
specialized providers and grocery stores because consumers perceive the meal-kits from both providers to be different (Nielsen, 2017; Moorhead, 2016). So, this study wants to empirically verify if the drivers for the consumption of meal-kits from both providers are different. This study specifically looks at price, effort reduction and healthiness as the drivers for the consumption of meal-kits because meal-kits from both providers differ the most from each other on these characteristics (Hertz and Halkier, 2017; Nielsen, 2017; Moorhead, 2016).
Furthermore, it is not known whether different levels of brand loyalty have an effect on the drivers price, effort reduction and healthiness. It is known that brand loyalty has implications for consumer behaviour (Morgan and Hunt, 1994; Zeithaml, 1994; Petty and Cacioppo, 1986a). Loyal customers behave differently than non-loyal customers do and tend to be more satisfied with and committed to a particular brand or provider (Morgan and Hunt, 1994). It is known that customers who buy a product once will better analyse the product and the implications it has for their personal health and are more critical regarding price and effort that is needed to consume the product (Petty and Cacioppo, 1986a). This is because their decision is made via central route processing were loyal customers often use peripheral route processing when making a repeat purchase (Petty and Cacioppo, 1986a). Resulting from this is that loyal customers are for example less price sensitive than non-loyal customers (Morgan and Hunt, 1994). Loyal customers stick with their brand or provider even when the price rises or a better alternative is introduced, which shows commitment (Morgan and Hunt, 1994). So loyal customers are in general less susceptible or motivated by price, effort or health
considerations when they want to purchase a product (Morgan and Hunt, 1994). With more actors entering the market of meal-kits, providers need to have a solid understanding what drives people to purchase meal-kits and if brand loyalty has an moderating effect on the relationship between price, effort reduction and healthiness on the purchase intention for meal-kits. If meal-kit providers do not have enough knowledge on the drivers and the effect that brand loyalty has on these drivers, they cannot optimize their marketing actions (Morgan and Hunt, 1994). This study wants to contribute to solving this problem by generating useful insights. In sum, this study measures to what extent the level of brand loyalty moderates the relationship between the drivers and the purchase intention of meal-kits.
1.5 Practical relevance
This study can provide providers of meal-kits with useful insights to create knowledge on the drivers for the purchase intention of their customers so that they can optimize their
actions accordingly and stimulate purchases (Brunner et al., 2010). The information generated by this study can help providers of meal-kits to develop marketing strategies that clearly distance meal-kits from traditional convenience foods which are in general perceived as being cheap and unhealthy (Friends and Goldin, 2019). Moreover, obtaining new customers is more expensive than retaining old ones (Venugopal and Pulidindi, 2015) and therefore providers of meal-kits must do their best to keep their customer satisfied and create loyalty wherever possible.
This study will generate insights on the difference between specialized providers and grocery stores which can help both providers in making allocation decisions and in shaping their strategies aimed at competitors. When it is found that certain drivers are more important for consumers of meal-kits from grocery stores compared to specialized providers, retailers such as AlbertHeijn can shape there strategies and tactics to meet those demands and be able to better compete with popular specialized providers. In turn, HelloFresh can use these insights to counter such strategies from grocery stores.
Furthermore, this study is aiming to give insight on the possible effect of brand loyalty on the drivers for meal-kit purchase intention. This is important because when for example AlbertHeijn introduces a new kind of meal-kit is it useful for them to have insight on how the level of brand loyalty consumers have towards their provider affect their behaviour and reaction to changing levels of price, effort reduction and healthiness. Not every consumer of meal-kits has the same level of loyalty regarding their brand. This research wants to test whether consumers with varying levels of brand loyalty have different drivers for their consumption. This information can help retailers and marketers get a better understanding of who their loyal customers are and how to keep them satisfied. These insights can also help providers to create loyalty programs to keep their loyal customers close and increase their spending. Both HelloFresh and AlbertHeijn can develop more effective marketing strategies to foster their loyal customers and attract new ones with this information.
1.6 Outline
The remainder of this thesis is structured as followed: The next part will form the theoretical basis which explains the variables central in this study and the relationship between these variables. Thereafter the methods that are applied in this study will be
discussed. Subsequently the analysis of the data is discussed follow by the results which ends in a conclusion. Thereafter a discussion of the results is carried out were the theoretical and
practical implications are discussed. To conclude, this study will explain its limitations and give possible future directions of research.
2. Theoretical framework
2.1 Introduction
This chapter will contain the theoretical background that forms the basis of this study. The key concepts will be defined and explained. The relations between the key concepts and the moderating variables will be explained according to the existing literature. The conceptual framework central to this study will be presented which will give a visual representation of the key concepts, moderating variables and their relationships. The expectations about possible outcomes will be presented by the formulation of hypotheses.
The next part will explain the theoretical background for this thesis. All variables will be discussed according to the existing literature and it will become clear what is meant by each variable in this specific study. The concepts discussed below are convenience food and its drivers, meal-kits, purchase intention and brand loyalty.
2.2 Convenience food
Convenience food is not clearly defined in the literature (Bech-Larsen and Grunert, 2003). But there is consensus that the purchasing and preparing of convenience foods involves saving time, mental and physical effort and in general making the process easier (Bech-Larsen and Grunert, 2003; Costa et al., 2001; Jackson and Viehoff, 2015).
Interest in convenience food started around 1920 and from there a steady stream of research emerged (Scholliers, 2015). Academic interest in the topic gradually increased with the growth of convenience foods that are introduced into the ‘real’ world. Convenience foods is a very broad concept encompassing different themes and the interest of academics shifted a couple of times from 1920 till 2015 (Scholliers, 2015). Scholliers (2015) identified four themes with corresponding time periods by analysing the google scholar database from 1920 till 2014. The first theme called ‘Production’ included keywords like manufacturing, retailing, technology and advertisement of convenience foods. This theme showed the novelty of the subject and dominated the literature from 1920 till 1960. The literature in this period
predominantly focused on how convenience foods were made, brought to the attention of the general public and sold to consumers. The second theme he identified is called ‘Choices’ and this theme included keywords like preferences, times, skills and quality. In this theme interest
was mainly focused on the reasons why convenience food was becoming more popular. This theme was dominant in the period from 1960 till 1980. The third theme is called ‘Households’ which focused on keywords like women, children, households, purchasing power and
expenditures. The main interest in this theme was aimed at uncovering the way in which the keywords above contributed to the growth and success of convenience foods. This theme was most studied in the period from 1990 till 2010. The fourth and final theme Scholliers (2015) identified is called ‘Health’ and contained keywords like obesity and health. This theme took off from the year 2000 when interest into the negative effects and poor nutritional value of convenience food increased. The themes mentioned above are not mutually exclusive and in some periods there was academic interest in multiple themes at the same time (Scholliers, 2015).
There has been a lot of debate on how to properly define convenience food (Jackson and Viehof, 2015; Costa et al., 2001; Marshall and Bell, 2003). But there is consensus in the literature that convenience food makes the process of consuming a meal easier (Jackson and Viehof, 2015). This can be manifested in terms of time-saving, cost-saving or reduced effort involved (Jackson and Viehof, 2015). A formal definition is given by Brunner et al. (2010; p. 499) they define convenience foods “as those that help consumers minimalize time as well as physical and mental effort required for food preparation, consumption and clean-up” The process of consuming a meal consist of preparation, planning, eating and cleaning up afterwards (Bech-Larsen and Grunert, 2003). In this study the definition by Brunner et al. (2010; p. 499) is used to understand convenience foods. Convenience foods are forms of nutrition in which one of the phases of consuming a meal needs reduced effort in comparison with the ‘normal’ way of consuming a meal. The ‘normal’ or traditional way of cooking a meal entails buying or obtaining the ingredients yourself and then plan and carry out the task that make up cooking a meal (Jackson and Viehof, 2015).
There are different categories of convenience foods and Marshall and Bell (2003) argue that these categories cannot be seen as distinct from another but rather as on a
continuum with convenience foods on the one side and traditional home-made meals on the other. The most convenient foods are meals from for example McDonalds or Burger King because with these meals the lowest amount of effort is needed to order and eat the meals they offer, often referred to as ‘fast-foods’ (Jackson and Viehoff, 2016). The least convenient meals are traditional home-made meals where a consumer must come up with a recipe, find the ingredients, plan, prepare and cook the meal themselves. The meal-kits central is this study are somewhere in the middle of the continuum between the most convenient foods
(fast-foods) and traditional home-made meals. This is because with meal-kits some tasks such as coming up with a recipe and finding the ingredients in the store are done for you by the producer, making it more convenient in comparison with traditional home-made meals. On the other hand, meal-kits still require the consumer to cook the meal themselves at home demanding more effort than ordering fast-food (Marshall & Bell, 2003).
In the literature on convenience food multiple drivers are identified as having an influence on the consumption and purchase intention of convenience food. In this study the drivers price, effort reduction and healthiness are chosen because these were the aspects were meal-kits differ the most from compared to traditional convenience foods (Nielsen, 2017; Moorhead, 2016). Furthermore, consumers perceived the meal-kits to be the most varying on these aspects between specialized providers and grocery stores.
The drivers price, effort reduction and healthiness apply to the functional and monetary dimensions that are mentioned above. Price is the quantity of payment or
compensation given by one party to another in return for one unit of goods or services. In this thesis only monetary payment is included. Price consists of actual (objective) price and the reference price. The actual price is the objective price that must be paid to obtain a meal-kit and the reference price is the price against which the actual price is compared by consumers. The actual price is seen relative to the perceived worth of a meal-kit. This means that
consumers compare the actual price with the product attributes and the price and attributes of a similar product to form the perceived value of a good, in this case meal-kits (Zeithaml, 1988). Previous studies indicated that price is an important factor in the decision making process regarding food consumption (Brunner et al, 2010; Hertz and Halkier, 2017; Steptoe et al., 1995; Zeithaml, 1988; Zeithaml et al., 2006; Lewis and Shoemaker, 1997). Price together with product features are the two most important decision variables used by marketers to influence the purchase behaviour of potential customers (Chang and Wildt, 1994). Price reflects the monetary dimension of perceived value and influences the purchase intention for meal-kits.
The second driver for convenience food is effort reduction. This is the amount of effort and time involved relative to the amount needed to make a ‘traditional’ meal. Meal-kits reduce effort because the recipe and the amounts of ingredients needed are specified. This way a consumer does not have to devote mental or physical effort to come up with or find a recipe online or weigh the right amount of ingredients (Brunner et al., 2010). Effort consist of mental and physical effort. Mental effort is the cognitive involvement needed to make a meal or think of making a meal (Hartmann et al., 2013). Thus, the time spent thinking about a
recipe or the right amount of ingredients for example is considered mental effort. Thinking about which provider to choose from or simply the mental stress that is caused by the thought of making a meal after a busy working day is considered mental effort. Physical effort, as the name suggests, is the amount of physical labour or effort that is needed to complete meal or meal-kit (Hartmann et al. 2013). The reduction of effort involved with cooking is being valued by many people who feel that they are under time pressure and therefore try to minimize time and effort spend on cooking a meal (Costa et al., 2007; Daniels and Glorieux, 2015). This is one of the main reasons for people to consume convenience foods (Brunner et al., 2010). Because meal-kits are a form of convenience food it plays a role as being a driver for the consumption of meal-kits (Brunner et al., 2010; Daniels and Glorieux, 2015).
The third and last driver is healthiness. Previous studies identified healthiness or the contribution of a meal to a healthy diet as an important determinant in the purchase decision of consumers (Brunner, 2010; Steptoe et al., 1995; Geeroms et al., 2008). Healthiness is a very broad term and there are multiple interpretations to what exactly constitutes healthiness (Geeroms et al., 2008). There is an important difference between personal health and the healthiness of a meal-kits in terms of nutritional value. Personal health refers to a person’s overall well-being and the healthiness of a meal refers to the nutritional value that a meal contains (Geeroms et al., 2008). Thus, in this study healthiness refers to the effect a meal-kit has on a human body and not the healthiness of the human body itself. In addition, healthiness only refers to nutritional value and vitamins it contains. Since the 1980’s the world health organization observed a rise in obesity and related diseases and brought this under the
attention of the general public. This resulted in people being more conscious about what they consume (Kozup et al. 2003). People became more concerned with the nutritional value a meal contains and began analysing it before purchasing and consuming a meal. Therefore, the increased attention to the healthiness of convenience foods resulted in manufactures
increasing the nutritional value and decreasing the amount of fat, sugar and salt that
convenience foods contained (Kozup et al. 2003) Nowadays, the healthiness of a meal is one of the most important drivers for the purchase and consumption of food (Brunner, 2010; Steptoe et al., 1995; Geeroms et al., 2008; Hoffman, 2016). People are aware of the often poor nutritional value and amount of vitamins convenience foods contain. Therefore, healthiness is a less stronger driver for convenience foods then for meal-kits, because consumers expect a meal-kit to be healthy (Kozup, 2003; Geeroms, 2008).
This study focusses on one specific form of convenience food namely meal-kits. Meal-kits are packaged meals containing the right amount of ingredients for a specified number of people and has a detailed instruction on how to make the meal (Nielsen, 2017). Because consuming a meal-kit takes away some effort of for example coming up with a recipe and shopping for the ingredients but still needs some effort in cooking the meal, it can be seen as a semi-convenience meal (Daniels and Glorieux, 2015). As explained in the introduction, the best way to look at the different categories of convenience food is that there exists a
continuum, according to Marshall and Bell (2003). Meal kits are then somewhere in the middle of this continuum and can be termed as semi-convenience foods (Daniels and Glorieux, 2015).
The introduction of meal-kits in grocery store is a reaction on the success of the specialized meal-kit providers. The concept of the meal-kits provided by two different producers is the same but the perceptions from consumers about both varies (Nielsen, 2017). Some authors argue that meal-kits from specialized producers are perceived to be of higher quality, more varied, healthier and give a better ‘eating-out’ experience (Nielsen, 2017; Moorhead, 2016). Furthermore, they argue that meal-kits from specialized providers are also perceived as more expensive and time consuming than meal-kits from grocery stores (Nielsen, 2017; Moorhead, 2016). The difference in perceptions of consumers about the meal-kits from specialized meal-kit providers versus those offered by grocery stores is not empirically verified (Hertz and Halkier, 2017). The specialized providers tend to offer meal-kits on a subscription base where you can determine yourself how often you get a meal-kit send to your home (HelloFresh, 2015). Initially HelloFresh started with a subscription based meal-kit delivery service but consumers are now also able to order on-demand and just once
(HelloFresh, 2015). When ordering a meal-kit from a specialized provider such as HelloFresh a consumer can choose the number of people the meal must serve and the number of meals the box must contain (HelloFresh, 2015). Thus, it is possible to specify the number of people the meals must feed and the number of different meals a consumer prefers. For example, a household consisting of four can order a meal-kit containing five different meals for Monday to Friday. This way the working parents do not have to put in effort coming up with recipes and buy the ingredients in the grocery store. The concept of meal-kits offered by grocery stores is the same as the meal-kits offered online by specialized providers (Nielsen, 2017; Hertz and Halkier, 2017). However, there are important differences regarding specific attributes such as price, ingredients, specific diets, nutritional value, production method
(organic/non-organic), preparation time and overall effort needed to make the meal (Hertz and Halkier, 2017). These attributes influence the perceived value of meal-kits for consumers (Taylor and Baker, 1994). In sum, meal-kits from specialized providers are more expensive, demand more effort and are healthier compared to meal-kits from grocery stores (Nielsen, 2017; Moorhead, 2016; Hertz and Halkier, 2017).
Brunner et al. (2010) conceptualizes convenience food in a broad sense making the distinction between highly processed foods, moderately processed foods, single components (canned beans, frozen fish sticks) and salads. Meal-kits cannot be attributed to one of these categories because meal-kits have elements from multiple categories. For example, a meal-kit often contains pre-cut vegetables or seasoned meat products. Those can be regarded as single components but also as moderately processed foods. Meal-kits therefore can be regarded as a separate and new category with providers being very creative and divers in the elements they include into their meal-kits (Nielsen, 2017).
2.2.2 Purchase intention
The perceived difference in meal-kits from specialized providers and grocery stores results in a perceived value difference (Kim and Chung, 2011). Many authors suggest that the perceived value of a product is influencing the purchase intention of consumers for that product. According to the theory of reasoned action, intention is the immediate antecedent of actual behaviour. (Fishbein and Ajzen, 1967). Thus, when meal-kit providers can uncover the customers’ intentions they have knowledge on what their behaviour will look like. The purchase intention of consumer is argued to be a reliable predictor of future sales and therefore a very useful construct for retailers (Ajzen, 1985; Kim and Chung, 2011). In this study purchase intention is defined as the probability that a consumer buys a meal-kit. This definition is based on the theory of planed behaviour developed by Ajzen (1985) the origin of this theory can be found in the theory of reasoned action first proposed by Fishbein and Ajzen (1967). According to the theory of reasoned action, if people evaluate the suggested behaviour as positive (attitude), and if they are convinced other people want them to perform a specific behaviour (subjective norm), this results in a higher intention (motivations) and they are more likely to behave in that way. A high correlation of attitudes and subjective norms to
behavioural intention, and subsequently to behaviour, has been confirmed in many studies. Sheppard et al. (1988) conducted a meta-analysis of past research and found confirmation of this high correlation in multiple studies. Intentions in general cause behaviour and the
expected behaviour of consumers is valuable for meal-kit providers because it can be linked to purchase intention and subsequently purchase behaviour (Kim and Chung, 2011). Previous studies have indicated that providers of meal-kits can benefit from an indication of future sales and revenue in terms of amount of ingredients or package material to buy and stock (Kim and Chung, 2011; Sniehotta, 2009).
An important antecedent of purchase intention is the perceived value of a good. The perceived value of a product by consumers is based on an evaluation of the benefits and costs of product when compared with other products. It is defined as ‘the consumer’s overall assessment of the utility of a product based on perceptions of what is received and what is given’ (Zeithaml, 1988; p. 14). Perceived value consists of the dimensions functional value, monetary value, social value and psychological value.
2.2.3 Brand loyalty
Loyalty is a very broad concept and there are two main research streams regarding brand loyalty. These are attitudinal loyalty and behavioural loyalty. This study looks at brand loyalty from an attitudinal perspective. Consumers that have attitudinal loyalty engage in extensive problem-solving behaviour through which they form brand preferences by comparing brands and their attributes (Bennet and Rundle-Thiele, 2002). The following definition of attitudinal loyalty is adopted from Berkowitz, Jacoby and Chestnut (1978; p. 89). Attitudinal loyalty is “The consumer’s predisposition towards a brand as a function of
psychological processes”. The psychological processes are attitudinal preference and commitment towards a brand (Jacoby and Chestnut, 1978).
The focus is on attitudinal loyalty because when a company want to design marketing programmes, they first have to understand attitudinal loyalty to be able to impact behavioural loyalty. Behavioural loyalty is the observable outcome of attitudinal loyalty reflected by for example market share or sales (Bennet and Rundle-Thiele, 2002). To ultimate goal of many marketers is to build a large loyal customer base. To reach this goal marketers must first develop an understanding of which customers have an attitude that results in potential brand loyalty (Bennet and Rundle-Thiele, 2002). Attitudinal loyalty can be understood on an individual level and on a brand level. The individual level is reflected by the propensity to be loyal. This means some people have a personality trait that causes them to stick with what they know and are loyal to the same provider in multiple product categories (Bennet and Rundle-Thiele, 2002). The brand level is reflected by the attitude towards the act of
purchasing a specific brand. This means some people have strong preferences for one brand or providers over another. The brand level of attitudinal loyalty is used to measure the concept in this study.
Brand loyalty indicates the extent to which customers are devoted to a company’s products or services and how strong their tendency is to select one brand over the competition (Morgan and Hunt, 1994). Loyal customers behave differently than non-loyal customer. They tend to be more satisfied with and committed to a specific brand or provider (Morgan and Hunt, 1994). Brand loyalty has some favourable implications for providers of meal-kits. Loyal customers have a strong desire to maintain the relationship with their provider. Furthermore, loyal customers are less critical in evaluating a specific offer against those of the competition (Morgan and Hunt, 1994). This is because interaction with a specific provider in the past resulted in positive customer experiences and overall satisfaction with the product and
relationship with the provider (Zeithaml, Berry and Parasuraman, 1994). Loyal customers are willing to pay extra or take an offer that does not precisely meet their demands because they perceive some unique value in the brand or provider that no alternative can meet (Zeithaml, Berry and Parasuraman, 1994).
It is known that customers who do not show any loyalty to one specific provider will compare the available providers and their offerings extensively before purchasing (Petty and Cacioppo, 1986a). This means they better analyse the different meal-kits and features regarding price, potential to reduce effort and health implications. This is because their
decision is made via central route processing where loyal customers often use peripheral route processing when making a purchase because they are already familiar with a provider and his offerings (Petty and Cacioppo, 1986a). Resulting from this is that loyal customer are less price sensitive, less health orientated and are prepared to put in some extra effort regarding meal-kits provided by their preferred brand compared with non-loyal customers (Morgan and Hunt, 1994).
2.3 Conceptual framework
Because meal-kits can be regarded as a relatively new and understudied category of convenience food it is relevant to verify and measure if the drivers of convenience food differ in their importance when compared to meal-kits. The drivers price, effort reduction and healthiness are important drivers for the consumption of convenience food in general and meal-kits differ from convenience food on these drivers (Brunner et al., 2010; Daniels and
Glorieux, 2015; Hertz and Halkier, 2017; Nielsen, 2017). Another reason to select price, effort reduction and healthiness out of numerous drivers that are important (Brunner et al. 2010), is that consumers perceived meal-kits to differ most from each other on those characteristics (Nielsen, 2017; Moorhead, 2016). The moderating effect of the provider of meal-kits is taken into account because the drivers for the consumption of the meal-kits varied between providers (Nielsen, 2017; Moorhead; 2016). The moderating effect of brand loyalty is included because the level of brand loyalty a customer has determines to what extent price, effort reduction and healthiness are important when making a purchase decision (Morgan and Hunt, 1994; Petty and Cacioppo, 1986a; Zeithaml et al., 1994).
(Figure 1. Conceptual model)
2.4 Hypotheses.
The next part will discuss the hypotheses that will be tested in this study. The hypotheses reflect the expectations regarding the relationships between the variables being studied. These relationships are visually presented in the conceptual model. The hypotheses are grounded in previous research and this literature will be discussed below.
2.4.1 Price Purchase intention for meal-kits Price Effort reduction Specialized providers versus grocery stores
Healthiness Control variables
- Familiarity with meal-kits
- Age - Gender
- Working status Brand Loyalty
Previous studies have indicated that price and quality are the two most important determinants for the perceived value of goods and therefore also meal-kits (Zeithaml, 1988; Zeithaml et al., 2006; Lewis and Shoemaker, 1997; Chang and Wildt, 1994; Dickson et al., 1985; Ahtola, 1984). Perceived value in turn determines to a large extent the purchase intention for convenience food. Price can be understood as an indicator of quality because in general a higher price indicates higher quality of a good. But this price-quality relationship only holds for products where the consumer can obtain little knowledge on product features and attributes (Chang and Wildt, 1994). The product attribute information for convenience foods is often clear and easy to find on the package. Therefore, when consumers purchase convenience foods, the price is not the most important factor in reflecting the quality of the food. People will form their perceived value more strongly relying on actual product attributes and for example reviews by other consumers. Therefore, consumers of convenience food are insensitive to price (Jensen and Yen, 1996). In other words, they are willing to pay more for convenient meals. This is because convenience foods decrease the amount of time and effort that is needed to complete a meal (Hertz and Halkier, 2017). Other factors such as an increase in income, more women participating in the labour force, changes in household demographics and changing lifestyles also cause consumers of convenience food to be less price sensitive (Jensen and Yen, 1996; Nayga, 1996; Prochaska and Schrimper, 1973). Meal-kits are a special form of convenience food because they still demand some effort and time to complete the meal. Consumers who buy meal-kits, value the aspect of time and effort saving, and at the same time want to experiment with new recipes and eat healthy (Geeroms, 2008; Moorhead, 2016; Nielsen, 2017). Therefore, consumers will better examine the options regarding the available meal-kits and all its attributes including price, by using central route processing (Petty and Cacioppo, 1986a). This causes consumers to be more price sensitive regarding meal-kits (Buckly, 2007; Hertz and Halkier, 2017).
An empirical study done by Chang and Wildt (1994) resulted in 414 valid responses to an experimental design. One of the relations tested was the relationship between price and purchase intention. The actual (objective) price was found having a significant negative effect on purchase intention in all settings (Chang and Wildt, 1994).
Based on these studies hypothesis 1 is formulated.
H1: Price has a negative effect on the purchase intention for meal-kits.
One of the trends prevalent in Western societies these days is the ‘lack of time’
principle. People increasingly feel that they do not have enough time to shop, prepare or make a meal at home (Costa et al., 2001; Costa et al., 2007). Furthermore, people increasingly feel that cooking at home is a bother and perceive cooking a meal at home as time-consuming. Therefore, people tend to look for ways in which the time and effort spend on consuming dinner is minimalized (Daniels and Glorieux, 2015; Costa et al., 2007). Meal-kits are a solution to the problem explained above. In general meal-kits reduce time and effort
compared to a ‘traditional’ meal cooked at home but not as much as ‘traditional’ convenience foods. Previous empirical studies have indicated that reduced effort is an important driver for the consumption of convenience food (Brunner et al., 2010; Hertz and Halkier, 2017; Daniels and Glorieux, 2015; Costa et al., 2007). One study used the food frequency questionnaire and found that a reduction in mental and physical effort as well as time-saving were significant drivers for the purchase behaviour and consumption of convenience food (Brunner et al., 2010). These findings are also applicable to meal-kits because meal-kits can be regarded as a distinct category of convenience foods. Therefore, the literature on convenience food is used to form expectations about meal-kits. Costa et al. (2007) found that Dutch citizens
increasingly made use of home replacement meals (HRM) because they are convenient and time saving. Meal-kits can be seen as an HRM. Another study done in the UK found that the average amount of time spent on cooking a meal was 8 min and that the average number of meals cooked from scratch was one meal a week (Buckley, 2007). This was because UK housemothers increasingly perceived to be under high time-pressure and made use of
convenience food to reduce effort (Buckley, 2007). Consistent with this literature hypothesis 2 is formulated.
H2: Effort reduction has a positive effect on the purchase intention for meal-kits.
2.4.3 Healthiness
The literature suggest that healthiness nowadays is one of the most important drivers for the purchase decision regarding food (Steptoe et al., 1995; Brunner et al., 2010; Geeroms et al., 2008). With a rise of obesity, diabetes and other health issues that can be attributed to foods with poor nutritional value, the public is becoming more conscious about their diets (Geeroms et al. 2008; Hoffman, 2016). Convenience foods already contain far less amounts of sugar, salt and saturated fat than twenty years ago (Hoffman, 2016) The growing awareness of the relationship between diet and overall health cause people to prefer healthy foods and even
choose healthiness over taste (Bogue et al., 2005). Convenience food is primarily consumed because it saves time and effort but people increasingly abstain from convenience foods because they are unhealthy (Steptoe et al., 1995; Geeroms, 2008). With meal-kits, the aspect of healthiness is more important for the purchase decision of consumers compared to
convenience foods (Moorhead, 2016; Nielsen, 2017). Healthiness is becoming a determining factor for many people in the choice for the convenience foods and meal-kits they consume (Geeroms et al., 2008). In the study conducted by Geeroms et al. (2008) a couple of health-related motives orientations (HRMO) that were found to be decisive in the choice for convenience food were identified. HRMO’s are abstract goals that motivate people to eat healthy and thereby becoming or staying healthy (Geeroms et al., 2008). They conducted an online survey with N = 1934 valid responses. This resulted in identification of several motivations for people to eat healthy. Physical and emotional well-being as well as energy and enjoying life were the most important motivations for people to choose healthy forms of convenience food over less healthier forms. In a study by Steptoe et al. (1996) the Food Choice Questionnaire identified nine categories that were important for the choice of food. Health and Weight control were found as having the biggest effect on the ultimate food choice of people (Steptoe et al., 1996). What is interesting is that this study proves that people prefer healthiness over for example price, convenience or familiarity. Meal-kits are considered as being healthy and convenient at the same time (Geeroms, 2008). For both convenience foods and meal-kits, an increase in the healthiness of the meal will increase the purchase intention for consumers (Geeroms, 2008; Steptoe, 1996; Brunner, 2010). Taken this empirical material into account one can conclude that people are more intended to purchase healthier food relative to foods that are less healthy. Therefore hypothesis 3 is formulated.
H3: Healthiness has a positive effect on the purchase intention for meal-kits.
2.4.4 Specialized providers versus grocery stores
Because of this difference in perception about the price, effort reduction and healthiness from both providers the drivers for the intention to purchase them potentially varies (Nielsen, 2017; Moorhead, 2016; Hertz and Halkier, 2017). An empirical study conducted by Hertz and Halkier (2017) showed that the parents as decision makers have a more positive perception of meal-kits offered by specialized providers compared to meal-kits from grocery stores (Hertz and Halkier, 2017).
The meal-kits that specialized providers offer are more expensive than those offered by the grocery stores competing with them (Hertz and Halkier, 2017). Partly, this has to do with the shipping costs that many providers include in the total price that consumers must pay (Hertz and Halkier, 2017; Moorhead, 2016). This also has to do with the fact that, on average, more specialized and expensive ingredients are being used in meal-kits from specialized providers. When segmenting the market for meal-kits Buckley et al. (2007) found that families that subscribe to a meal-kit delivery service were mostly middle- and upper-class families with an above average disposable income. Those families are less price sensitive than families with an average or below average disposable income. (Buckley, 2007). Because meal-kits from specialized providers already being perceived as more expensive than those from grocery stores the same increase in price will result in a greater decline in purchase intention for meal-kit from grocery stores than for specialized providers. This is because families that buy meal-kits in the grocery store are more price sensitive (Buckley et al., 2007; Zeithaml et al., 2006; Lewis and Shoemaker, 1997). Consumers that buy meal-kits in the grocery store are more price sensitive because there are more available substitutes to choose from. Instead of buying a meal-kit, consumers can find and buy all ingredients separately and by doing so save money (Lewis and Shoemaker, 1997). Furthermore, consumers that shop in the grocery store have a better reference price than consumers that buy meal-kits from
specialized providers (Zeithaml et al., 2006). This is because when consumer consider buying a meal-kit in the grocery store they will compare the price of the meal-kit to the price they last paid or the average price they spend on buying a meal in the grocery store (Zeithaml et al., 2006). With specialized providers, the meal-kits cost roughly the same and meal-kits of specialized providers are already perceived as more expensive as compared to those of grocery stores and therefore the reference price is also higher in the mind of the consumer (Zeithaml et al., 2006). Moreover, mostly middle- and higher-income families use specialized providers and their price limits are higher than those of the average consumer that buys kits in the grocery stores (Zeithaml et al., 2006). Price limits means that if a price of a meal-kits falls below the lower limit consumers may doubt the quality and when it is above the upper limit it will be judged as too expensive (Zeithaml et al., 2006). Taken this literature into consideration, hypothesis 4a is developed.
H4a: Price has a greater negative effect on the purchase intention for meal-kits from grocery stores than for specialized providers.
The meal-kits offered by specialized providers demand more effort than meal-kits from grocery stores (Nielsen, 2017; Moorhead, 2016). This has multiple reasons. Specialized providers offer a broader assortment of meal-kits making the choice demanding more mental effort when compared with just three options in your local grocery store (Hertz and Halkier, 2017). Furthermore, specialized providers regularly offer recipes that are not very traditional. This is because customers from specialized providers prefer to be inspired with recipes they have never encountered before because this gives them an eating-out experience (Hertz and Halkier, 2017). So, in general the motivation for purchasing a meal-kit from a specialized provider is more strongly to experiment and the motivation for purchasing a meal-kit from a grocery store is more to get a quick meal (Hertz and Halkier, 2017). Because consumers are not familiar with the recipes and kitchen tools to use, meal-kits from specialized providers demand more effort than grocery store meal-kits. In addition, meal-kits from specialized providers are found to be more time-consuming than those from grocery stores. In a comparison of meal-kits from the five most popular brands in the Netherlands including HelloFresh and the Allerhande box from AlbertHeijn it was found that a box from the specialized providers took five a ten minutes longer on average to complete the meal (‘Maaltijdboxen Vergelijken; Welke past bij jou?’, 2020). Meal-kits from specialized providers are on average more effort demanding, but consumers are motivated by the fact of experimenting with new recipes and therefore do not care much about the extra effort
(Nielsen, 2017). Consumer that buy meal-kits in grocery stores are stronger motivated by the effort reduction aspect, thus a reduction in effort will increase their purchase intention more than those of people that buy meal-kits from specialized providers (Moorhead, 2016). Therefore, hypothesis 4b is formulated.
H4b: Effort reduction has a greater positive effect on the purchase intention for meal-kits from grocery stores than for specialized providers.
Meal-kits from specialized providers are perceived to be healthier than those from grocery stores (Nielsen, 2017; Moorhead, 2016). The positive effect of an increase in
healthiness for a meal kit from specialized providers is larger than for meal-kits from grocery stores. Consumers that buy meal-kits from grocery stores are motivated to get a quick meal and value the aspect of healthiness less than consumers that buy meal-kits from specialized providers (Bogue et al., 2005; Geeroms, 2008). The potential increase in purchase intention is greater for specialized providers because consumers from both providers have different reasons and motivations to make use of a specific provider (Hertz and Halkier, 2017).
Consumers who buy meal-kits from specialized providers are more health orientated and are willing to devote some extra effort to cooking a healthy meal (Nielsen, 2017; Geeroms, 2008). Consumer who buy meal-kits from grocery stores are less willing to put in extra effort to cook healthy. Therefore, healthiness is more important for consumers that buy meal-kits from specialized providers compared to grocery stores (Hertz and Halkier, 2017).
Therefore, hypothesis 4c is formulated.
H4c: Healthiness has a greater positive effect on the purchase intention for meal-kits from specialized providers than for grocery stores.
2.4.5 Brand loyalty
Previous studies have indicated that brand loyalty has a moderating effect on the drivers for consumption of convenience food. These studies indicated that loyal customers are less price sensitive, less health orientated and willing to put in some extra effort (Morgan and Hunt, 1994). Loyal customers are in general sticking with their brand and provider even when the price rises or a better alternative is introduced, which shows commitment (Morgan and Hunt, 1994). When customers are loyal to a provider of meal-kits they analyse the product less extensive before purchase and use peripheral route processing to come to a decision. This means that price, effort reduction and healthiness are less important for loyal customers because they strongly prefer one provider of meal-kits over another. Brand loyalty causes consumers to stick with their brand or provider even if there exist a better alternative in terms of product attributes such as price, effort reduction or healthiness of a meal-kit (Zeithaml, Berry and Parasuraman, 1996). Loyal customers are in other words more resistant to counter persuasion than non-loyal customers are (Dick and Basu, 1994). This means that non-loyal customers are prone to persuasion by competitors with offerings that have better price, effort reduction or healthiness levels. So, loyalty towards a specific meal-kit provider results in customer valuing the relationship they have with a provider so much that attributes like price, effort reduction or healthiness become subordinate. Sometimes a provider rewards this loyalty by using loyalty programs or giving discounts for repeat purchasing (McIntosh and Lockshin, 1997). In a study by (Bloemer and Kasper, 1995) loyalty towards a brand or provider was found to be causing consumers to make repeat purchases with the same provider even if a competitor had better alternatives. Furthermore, the effect of price and other product attributes on purchase intention was found to diminish with increasing loyalty towards a specific brand
or provider (Bloemer and Kasper, 1995). Consistent with this literature hypothesis 5a, 5b and 5c are formulated.
H5a: Brand loyalty weakens the relationship between price and the purchase intention for meal-kits.
H5b: Brand loyalty weakens the relationship between effort reduction and the purchase intention for meal-kits.
H5c: Brand loyalty weakens the relationship between healthiness and the purchase intention for meal-kits.
3. Method
3.1 IntroductionThis study is designed to generate insights on the drivers for the purchase intention for meal-kits while accounting for the possible moderating effect of brand loyalty and the
difference between specialized providers and grocery stores. The data that is used is generated by a survey specifically developed for this study.
The following part will explain the methods applied in this research. The research design will be explained by elaborating on the motivation for choosing quantitative research methods and the methods used to collect data. Thereafter, it will be discussed how the variables central in this study are operationalized according to which scales are used. The final section of this chapter will explain the analysis and which tools are used to generate insights.
3.2 Research design
This study is quantitative in nature. The choice for a quantitative research design is based on the fact that this study is aimed at giving a generalizable answer on the central question. A quantitative research design is better suited to generate generalizable results (Vennix, 2000; Field, 2014). Qualitative research based on interviews or observations will need a large sample to generate generalizable insights and this simply is to time and labour intensive for the subject of this master thesis. Quantitative research is described by Vennix (2000) as the collecting of data from a relatively large number of comparable research objects and subsequently subjecting this data to statistical analyses to come to conclusions. Within quantitative research there are different methods to collect data. The method that is used and validated the most is the survey (Vennix, 2000; Field, 2014). This method uses a
questionnaire to collect data on opinions from a set of comparable research objects.
The data that is used in this study is collected through an online survey. The choice to use a survey in this research is based on multiple reasons. First, a survey enables a researcher to collect data from a large sample of respondents by handing out paper surveys or sending the surveys via e-mail or other online tools. When the sampling is done at random it is possible to generate insights which can be generalized to the total population of research objects. Second, a survey is the most cost-effective way to reach a lot of respondents. Third, a
survey ensures objectivity of the data because the respondents or their answers are not being influenced by the interpretation of an interviewer or subjective observations. Fourth, the questions in a survey are standardized so the answers will be unambiguous and easily interpretable. Finally, the data collected by a survey can be subjected to deeper statistical analysis with programmes such as SPSS which enables a researcher to make subgroups and uncover relationships between variables (Vennix, 2000; Field, 2014).
3.3 Procedure
The data is collected from a sample of the total population of Dutch meal-kits
consumers. The sample is selected at random, so it is possible to generalize the insights from the sample to the total population of Dutch meal-kit consumers (Vennix, 2000; Field, 2014). Furthermore, the survey is written in the Dutch language and only distributed to people who are Dutch. Therefore, all respondents are native Dutch speakers. This research uses a survey to collect data which will be analysed using the statistical data analysis program SPSS. The sampling is based on consumers who are in the age group 18 to 70. The lower bound of this age group is chosen because children below 18 rarely are the ones responsible for coming up with and making dinner in a household (Costa et al., 2007). The upper bound of this age group is based on the fact that people above 70 have generally routinized their food choice so are less willing to try something new such as the concept of meal-kits (van Deursen and Helsper, 2015). In addition, people above 70 are in general not very familiar with the internet and therefore not likely to order meals from one of the specialized providers who offer their meal-kits online. (van Deursen and Helsper, 2015). Vennix (2000) states that the minimal number of respondents is 20 to 25 per variable used in the conceptual model. As the conceptual model central in this research shows, there are three independent variables, and two moderating variables. This means the minimal number of respondents needed to generate meaningful results is 5 × 25 = 125.
3.3.1 Survey
The questionnaire that is used to collect data is specially developed for this study. The questionnaire is made with the program Qualtrics. This is an online tool that helps developing and distributing the questionnaire online. This research compares differences between two groups of respondents. This is called a between subjects design (Hair, Black, Babin, and
Anderson, 2014, pp. 352–368). This design means that the total respondent set is split into two groups of approximately the same size. The respondents are randomly assigned to one of the two groups using Qualtrics.
The choice for a between subjects design is based the fact that this study is interested in the purchase intention for meal-kits from HelloFresh and AlbertHeijn. Thus, half of the respondents is manipulated by seeing a meal-kit from HelloFresh. The other half is
manipulated by seeing a meal-kit from AlbertHeijn. After both manipulations, the two groups are asked about their purchase intention. Both groups get the same questions about their consumer characteristics and level of brand loyalty in the remainder of their questionnaires. Two separate groups are needed to be able to compare the purchase intention after the manipulation of both groups in a valid way. This design is needed to counter learning and priming by respondents. Because, when a respondent is confronted with both manipulations they can learn, form expectations, or develop other forms of cognitive biases that affects the validity of their answers (Charness, Gneezy and Kuhn, 2012). Furthermore, between subjects designs have shorter questionnaires compared to within subjects designs. Shorter
questionnaires are less tiring or boring compared with long questionnaires so there is less chance of response biases (Charness et al., 2012).
To minimize the number of missing values Qualtrics offers the option to force a response when answering the questions. This option is used, and this means respondents can only proceed to the next question if the previous one has a valid answer. In the survey the respondents are routed to either the HelloFresh manipulation or the AlbertHeijn manipulation. Because one independent variable is price and this study is partly interested in how price affects the purchase intention for meal-kits price is added below the picture of both
manipulations. This price is the actual price that must be paid to obtain such meal-kits in real-life. So, below the meal-kit from HelloFresh you can see that this kit contains three meals for four persons and costs € 61,95. The meal-kit from AlbertHeijn also contains three meals for four persons but costs € 51,50. These prices are based on the websites of both providers and on a website that compares the five most popular meal-kits in the Netherlands on aspects such as price (‘Maaltijdboxen Vergelijken; Welke past bij jou?’, 2020).
This study did not use a control group which is manipulated to act as a baseline to compare the results with. The reason for this choice is that the results can easily be compared to the existing literature, so no control group is needed (Vennix, 2000). A control group is most needed for study’s were the effect of the manipulation is unknown. In this study the effect is strongly based on the extensive literature review and therefore not totally unknown.
Before distributing the survey, a pre-test was conducted to check for inconsistencies and whether the questions were clear or not. The survey was pre-tested by five people and some minor changes were needed. The goal of the survey and all the questions proved to be clear. There were some suggestions. First, three people suggested that the introduction is a bit too lengthy, so it was shortened. Second, some people suggested that the part were the
consumer characteristic ‘inspanning’ is questioned also was a bit too long with fourteen items compared with the other characteristics only having five or six. Therefore, the items were reduced to ten. Some items had a lot in common, so deleting these items is not problematic for the validity of the measure. For healthiness one item was deleted because the pre-test showed that people found that item not appropriate in the context of meal-kits. Lastly, some spelling errors were resolved. The average time that was needed for the pre-test was eight minutes.
3.4 Research ethics
This research is conducted with important ethical considerations in mind. One crucial aspect of every research is that it does not cross any ethical boundaries (Goodwin, Pope, Mort and Smith, 2003). One of the most important rules is to carefully handle personal data
collected from respondents. Privacy must be considered with every decision. To ensure this privacy respondents who take part in the survey do this completely anonymously. The personal information like age, gender and education will only be used for research purposes and not made public. The respondents who take part do this completely voluntarily and are not obliged to finish the survey when the want to stop. It is assured that respondents who complete the survey have the possibility to look at the results of this research when it is finished.
It is also crucial to conduct research according to the code of conduct regarding academic writing. In the ‘Nederlandse gedragscode wetenschappelijke integriteit’ (2018) the rules for proper academic writing are explained. The rules in this document are developed to ensure academic integrity. This research is conducted according to these rules.
3.5 Measures
The variables central in this study are measured according to different scales. For price, effort reduction, healthiness, brand loyalty and purchase intention the scales are already
validated and proven in other studies. For specialized versus grocery stores two scenarios are developed. To see the exact scales and items used to measure the constructs see Table 1.
3.5.1 Purchase intention
The scale used to measure purchase intention is adapted from Ajzen and Fishbein (1980). The original scale consisted of three items measuring purchase intention. This scale is developed in the context of the theory of planned behaviour.
3.5.2 Specialized provider versus grocery stores
This research is interested in meal-kits from two kinds of providers, namely specialized providers and grocery stores. The specialized provider central in this study is HelloFresh. The choice for HelloFresh is based on the fact that the most popular specialized meal-kit provider in the Netherlands is HelloFresh (Nielsen, 2017; Moorhead, 2016;
SecondMeasure, 2017). HelloFresh started in Germany and soon delivered meal-kits across Europe. The success of HelloFresh resulted in many nationally oriented meal-kit providers to copy the concept of HelloFresh (SecondMeasure, 2017). HelloFresh is regarded by many as the leader in the meal-kit delivery market so it is useful to focus on HelloFresh in this study because results and insights will have implications for other specialized providers who operate using a comparable business model.
The Allerhande meal-kit from AlbertHeijn will be the representative for the meal-kits from grocery stores. The choice for the Allerhande meal-kit is based on the fact that
AlbertHeijn has been the most popular grocery store in the Netherlands over the past 10 years (Meijsen, 2017). Also, and perhaps related, is the fact that the Allerhande meal-kits is the most popular meal-kit from a grocery store in the Netherlands (Meijsen, 2017). Lidl and Jumbo are the second and third most popular grocery stores in the Netherlands. These grocery stores also have their own meal-kits, but the concept is the same. The meal-kits from
AlbertHeijn, Lidl and Jumbo are comparable on aspects such as price, effort needed to complete and healthiness (Meijsen, 2017). Therefore, the Allerhande meal-kit from AlbertHeijn is representative for the meal-kits from Lidl and Jumbo. This ensures that the results and insights regarding the Allerhande meal-kit will also have implications for the meal-kits from other grocery stores such as Lidl and Jumbo.
